Related Experiment Video
Updated: Aug 16, 2026
Profiling Maternal Behavior Responses During Whole-Brain Imaging
Published on: January 24, 2025
Neuroanatomical profile of externalizing problems level in adults: a machine learning study
Shaohuai Chen1, Yusong Zhang1, Jihang Zheng1
1Department of Neurosurgery, The Second Affiliated Hospital and Yuying Children's Hospital of Wenzhou Medical University, Wenzhou 325027, Zhejiang, China; Wenzhou Municipal Key Laboratory of Neurodevelopmental Pathology and Physiology,Wenzhou Medical University, Wenzhou, 325035, China.
Machine learning identified a brain structure pattern predicting externalizing problems (EXT) in young adults. This neuroanatomical signature, involving frontal-limbic circuits, offers potential for monitoring and prevention strategies.
Area of Science:
- Neuroscience
- Psychiatry
- Machine Learning
Background:
- Externalizing problems (EXT) are externalized behaviors with emerging neurobiological correlates identified via structural magnetic resonance imaging (sMRI).
- Generalizability and specific contributions of cortical thickness and surface area to EXT remain unclear.
Purpose of the Study:
- To characterize the neuroanatomical signature of EXT using machine learning.
- To assess the generalizability of neuroanatomical findings for EXT in a large, out-of-sample dataset.
Main Methods:
- Utilized the Human Connectome Project Young Adult dataset (N=1098).
- Employed a machine learning cross-validated elastic net regression approach.
- Validated findings using a held-out test set and univariate linear mixed effects modeling.
Main Results:
- A multi-region neuroanatomical signature robustly predicted EXT.
- The predictive model achieved R² of 2.69% (morphometry-only) and 6.45% (morphometry + demographics).
- The identified neuroanatomical pattern implicated regions within the frontal-limbic circuit.
Conclusions:
- Machine learning successfully derived a neuroanatomical profile for predicting EXT in healthy adults.
- Findings support the role of the frontal-limbic circuit in externalizing behaviors.
- The derived profile may inform future monitoring and prevention strategies for EXT in community populations.
